This Project, synergistic with Project 1, will evaluate the association between the gut microbiota and obesity and its associated metabolic phenotypes. Background: In work by Project 1, when fecal microbiota from obese (Ob) and lean (Ln) twins from discordant pairs (not metabolically phenotyped) were transplanted to germ-free mice, Ob microbiota transmitted increased adiposity and metabolic features that accompany human insulin- resistant states (e.g., increased circulating levels of serum branched-chain amino acids). In addition, co- housing mice harboring Ob co-twin microbiota with mice containing transplanted Ln microbiota prevented development of these body composition and metabolic phenotypes in the Ob cagemate: prevention was associated with unidirectional invasion of bacterial taxa from the Ln to the Ob cagemate's gut community, indicative of unfilled niches in Ob microbiota that could be occupied, but only under certain dietary contexts (observed under a low saturated fat/high fruit & vegetable [LoSF-HiFV) diet, but not under a HiSF-LoFV diet). Results were consistent with previous findings of reduced diversity in the microbiota of Ob individuals and suggest that fecal transplants from discordant twin pairs to gnotobiotic mice could help guide development of therapeutic strategies (microbial or diet-based) for restoration of the gut microbiota of the microbiota donor population to a pre-obese or metabolically healthy state. Approach: To better define the contributions of the microbiota to obesity and its associated metabolic dysfunctions, and to advance development of microbiota- directed interventions for treatment or prevention, we will identify, from a large panel of young adult female dizygotic twin pairs, three informative groups: pairs dually discordant for obesity and metabolically healthy/unhealthy (MH/MUN: indexed by hyperinsulinemic/euglycemic clamp [HEC]), pairs singly discordant for obesity, and pairs singly discordant for metabolic health (LnMH/ObMUN, LnMH/ObMH and ObMH/ObMUN pairs, Ln=BMI 18.5-24.9 kg/m2, Ob=BMI 30-44.9). Pairs will undergo HEC and body-composition analyses to quantify insulin-resistance and confirm group assignment. Pairs will provide fecal, urine and blood (serum) samples under free diet (1 week), with daily recording of diet intake, then complete an in-home cross-over controlled diet study, eating in random order a controlled HiSF-LoFV diet (2 weeks), and a controlled more healthy diet (LoSF-HiFV, 2 weeks), with continuing provision of fecal, urine, and end-of-diet blood samples. Human-to- mouse fecal transplant experiments (Project 1), using donor feces collected under LoSF-HiFV versus HiSF- LoFV diet conditions, will be used to characterize transmissibility of donor phenotypes, their sensitivity to diet and the degree of prevention of transmitted ObMUN and ObMH phenotypes achievable with co-housing with Ln mice (as a function of bacterial invasion pattern and diet type). Metabolomic analyses (Core A) will be used to test whether metabolic phenotypes observed in mice fed diets similar in composition to the human diets, represent those present in the humans. Increased recruitment of twin pairs will ensure adequate power.